Nouriel Roubini and Preston Byrne

The bitcoin blockchain dream is alive and well. Bitcoin blockchain dreams have led to predictions that bitcoin and other cryptocurrencies will fail typically elicit a broader defence of the underlying blockchain technology. Yes, the argument goes, more than half of “initial coin offerings” to date have already failed, and most of the 1,500-plus cryptocurrencies also will fail, but blockchain will nonetheless revolutionise finance and human interactions generally.

In reality, blockchain is one of the most overhyped technologies ever. For starters, blockchains are less efficient than existing databases. When someone says they are running something on a blockchain, what they usually mean is that they are running one instance of a software application that is replicated across many other devices.

The required storage space and computational power is substantially greater, and the latency higher, than in the case of a centralised application. Blockchains that incorporate “proof-of-stake” or “zero-knowledge” technologies require that all transactions be verified cryptographically, which slows them down. Blockchains that use “proof-of-work”, as many popular cryptocurrencies do, raise yet another problem: they require a huge amount of raw energy to secure them. This explains why bitcoin “mining” operations in Iceland are on track to consume more energythis year than all Icelandic households combined.

Blockchains can make sense in cases where the speed/verifiability trade-off is actually worth it, but this is rarely how the technology is marketed. Blockchain investment propositions routinely make wild promises to overthrow entire industries, such as cloud computing, without acknowledging the technology’s obvious limitations.

Consider the many schemes that rest on the claim that blockchains are a distributed, universal world computer. That claim assumes that banks, which already use efficient systems to process millions of transactions per day, have reason to migrate to a markedly slower and less efficient single cryptocurrency.

This contradicts everything we know about the financial industry’s use of software. Financial institutions, particularly those engaged in algorithmic trading, need fast and efficient transaction processing. For their purposes, a single globally distributed blockchain such as Ethereum would never be useful.

Another false assumption is that blockchain represents something akin to a new universal protocol, like TCP-IP or HTML were for the internet. Such claims imply that this or that blockchain will serve as the basis for most of the world’s transactions and communications in the future. Again, this makes little sense when one considers how blockchains actually work. For one thing, blockchains rely on protocols like TCP-IP, so it isn’t clear how they would ever serve as a replacement.